Breast cancer is the second leading cause of cancer deaths among American women, and thus additional, effective treatment protocols are needed. Lysyl oxidase (LOX) is an essential extracellular enzyme that controls matrix deposition. The LOX gene was isolated as the ras recision gene (rrg) with an ability to inhibit the transforming activity of the Ras oncogene in NIH 3T3 cells. Consistently many cancers display reduced LOX gene expression. The LOX gene encodes a 50 kDa pro-enzyme (termed Pro-LOX), which is secreted into the extracellular environment where it is processed by proteolytic cleavage to an 18 kDa amino terminal propeptide (LOX-PP) and a functional 32 kDa enzyme. The PI's group demonstrated that the rrg activity resides in the LOX-PP domain, whereas, the LOX enzyme itself has been found to promote a more aggressive phenotype and tumor metastasis under hypoxic conditions. Importantly, LOX-PP activity was shown by the PI's group to effectively inhibit breast cancer cells, driven by Her-2/neu which signals via Ras. In culture, LOX-PP suppressed the PI3K/Akt and MEK/Erk pathways and NF-?B. Oncogenic Her-2/neu induces epithelial to mesenchymal transition (EMT) and this was reverted by LOX-PP. LOX-PP induced E-cadherin, decreased levels of Snail, tumor migration and formation of branching colonies in Matrigel, and reduced tumor formation in nude mice. A single nucleotide polymorphism (SNP) (rs1800449) G473A, resulting in an Arg158Gln substitution in a highly conserved region in the LOX-PP domain, occurs with an average 24.6% 473A minor allele carrier frequency in the HapMap database. However when cancer cells were examined, the G473A SNP was present in the majority of lines and the LOX-PP Gln variant displayed substantially impaired tumor suppressor function in vitro and in vivo, and a reduced ability to oppose the pro-tumorigenic effects of LOX. In a pilot study of African-American women, the minor 473A allele was associated with increased risk of ER1 negative breast cancer. These findings lead to two related central hypotheses: 1. Wildtype LOX-PP represents an important tumor suppressor, the activity of which is compromised by an Arg to Gln substitution. 2. A G473A (rs1800449) SNP in the LOX gene leads to increased risk of more invasive cancers. In this new RO1 application, we propose to: (Aim 1) Elucidate the mechanisms whereby wildtype LOX-PP mediates its action as a suppressor of Ras signaling in breast cancer cells and determine how these are affected by the Arg158Gln substitution; (Aim 2) Test the ability of LOX-PP to impede the tumor promoting effects of the LOX enzyme in culture and in an orthotopic mammary fat pad model; (Aim 3) Test the hypothesis that the rs1800449 SNP represents a risk factor for more invasive breast cancer using a knock-in mouse that will be prepared. The proposed studies represent pre-clinical testing of LOX-PP as an inhibitor of Ras signaling and as a risk factor of disease progression. As the LOX-PP activity was also effective against lung and pancreatic cancers, our findings have broader implications for carcinomas driven by an activated Ras oncogene.